Genetic Variability of Two Populations of the Ectomycorrhizal Fungus Xerocomus Chrysenteron Associated with European Beech (Fagus Sylvatica L.)

GENETIC VARIABILITY OF TWO POPULATIONS OF THE ECTOMYCORRHIZAL FUNGUS XEROCOMUS CHRYSENTERON ASSOCIATED WITH EUROPEAN BEECH (FAGUS SYLVATICA L.)

Constanze von Meltzer & Gunter M. Rothe

Institut fiir Allgemeine Botanik, Fachbereich Biologie, Johannes Gutenberg-Universitat, D-55099 Mainz, Germany Correspondence: Dr. Gunter M, Rothe, Department of General Botany, Faculty of Biology, Johannes Gutenberg-University, D-55099 Mainz, Germany (tel.: +49-613 1-3922535; Fax: +49-6131-3925661; email: [email protected])

Received August 16, 1999; accepted February 2, 2000

ABSTRACT

The intra- and interpopulation genetic variability of the ectomycorrhizal fungus Xerocornus chrysenterorz (Bull. Ex. St. Amans) QuC1, infecting European beech (Fagus sylvatica L.), was studied in two populations in central Germany (Bingen and Heppenheim). The allele and genotype frequencies of the polymorphic enzyme loci Acp-B, Dia-B, Dia-C, G6pdll-A, Lap-B, Mdll-C, Me-B, Pepl-B and Pep2-B served as genetic markers. The total number of alleles of the nine gene loci was 23 (allele/polymorphic locus = 2.4). The alleles 1 of the loci Me-B and Din-B were observed only in one of the populations (Heppenheim). The average number of alleles per polymorphic locus was a little higher in one stand (Heppenheim: 2.55) than in the other (Bingen: 2.33). The number of effective alleles was calculated to be n, = 1.56 (Heppenheim) and n, = 1.53 (Bingen) in both populations. The genotypes Dia-B 111, G6pdh-A 112, Lap-B 113, MdhC 112 and Me-B 111,112, 113 were found only in one population (Heppenheim), while they were missing in Bingen. On the other hand, the Bingen population showed the genotypes G6pdh-A 111 and Lap-B 111 which were missing in the Heppenheim population. Relative allelic frequencies varied little (3 to 5 %) at the loci Dia-C, G6pdIz-A, Lap-B, Mdh-C, Pepl-B and Pep2-B whereas the frequency differences of the alleles 1 and 2 at the locus Acp-B were considerable in both populations (allele 1: 46 % (Heppenheim), 82 % (Bingen)). As a consequence, a low genetic distance (D = 3 %, do = 0.9 %) was observed among the two populations studied. Also, the average heterozygosity was not significantly different in both populations (HE= 33 76). An excess of heterozygous genotypes was observed in both populations at the loci Me-B and Pep2-B. The total genetic diversity was H, = 0.3466, the genetic diversity within populations was H, = 0.3381 and the amount of genetic differentiation of the total diversity was G,,= 2.5 %. These data are compared with known population genetic data of European beech and similarities and differences are discussed.

Key words: Basidiomycetes, ectomycorrhiza, European beech, Fagus sylvatica, isozymes, population genetics, symbiosis, Xerocornus clzrysenterori

INTRODUCTION However, the genetic basis of such intraspecific variation has not been investigated. So far, the ectomy- European beech (Fagus sylvatica L.) lives in close corrhizal fungus Suillz~stomentosus (Kauffm.) Singer, symbiosis with some 30 root ectomycorrhizal fungi Sue11 & Dick, occurring in the boreal forests of Canada, (AGERER1988- BRAND 1989,1991).The genetic is the only one in which the intra- and interpopulation variability of beech populations has been investigated genetic variability has been studied, using cluster intensively (MULLER-STARCK& STARKE1993; KON- analysis (ZHUeta/. 1988). In association with different NERT 1995; LEONARDI& MENOZZI1995; LOCHELT& tree species, the interpopulation genetic variability of FRANKE1995; HATTEMER&ZIEHE 1996), but the intra- the fungus was larger than its intrapopulation genetic and interpopulation genetic variability of their fungal variability (ZHUet al. 1988). partners is completely unknown. From other myco- The aim of this investigation was to compare the rrhizal fungi, it is known that different isolates of a inter- and intrapopulation genetic variability of the particular species can exhibit great variation in ecologi- ectomycorrhizal fungus Xerocomus chrysenteron (Bull. cal and physiological functions, such as in the synthesis Ex. St. Amans) QuCl and to relate the data obtained to of enzymes (LUNDEBERG1970), in the uptake of the genetic structure of its host European beech (Fagus mineral nutrients (LITTKE et a/. 1984), and in the sylvatica L.). symbiotic effectiveness (MOLINA1979; MARX1981).

O ARBORA PUBLISHERS C. VON MELTZER& G. ROTHE:GENETIC VARWTION OF ~EROCOMUSCHRYSENTERON

MATERIAL AND METHODS magnification according to the morphological criteria described by BRAND(1989) ~~~AGERER(1988-1998). Stand characteristics The very same clusters of one sample were then sepa- rated into root tips infected with the dikaryotic X. Fine roots (d s 2 mm) which were infected at their tips chrysenteron and non-infected proximal parts (d = 1 to with Xeroconzus chrysenteron (Bull. Ex. St. Amans) 2 mm) (cf. DAHNEet al. 1995). Mycorrhizal root QuCl were sampled in two beech (Fagus sylvatica L.) treelets having a fresh weight of at least 50 mg were put stands in Central Germany. in a 1.5 ml Eppendorf-safe-lock-tube, frozen for 10 sec One of the beech stands is seven hectars large and in liquid nitrogen and stored at -80 "C. Non-mycorrhi- located at the Odenwald hills, about 36 krn south of zal roots of the same root cluster were treated sirni- Frankfurt at Main, in the forest district of Heppenheim, larily. logging area Kirschhausen, compartment 101. The beech trees are approximately 100 years old (1999). Enzyme extraction They grow on a moderate to steep, north-west inclined, slope at 305 to 350 m above sea level on a loess loam, A 50 mg quantity of either mycorrhizal or non- above granite. The moderately well drained site was not mycorrhizal roots was homogenized with 100 pl limed. The last selective logging was performed in extraction medium and 7.5 mg of wet PVPP (polyclar winter 1994195. AT) as described by FIEDLERand ROTHE (1999). The other beech stand is eight hectars large and Afterwards, the slurry was thoroughly mixed with a located at the Hunsriick hills, about 45 krn west of spatula and stored on ice for 10 min. The extraction Frankfurt at Main, in the forest district of Bingen, medium consisted of (mg.100 ml-') of 100 mM Na- logging area Jagerhaus, compartment 56. The approxi- phosphate buffer, pH 7.0): cysteine (30), 2-mercapto- mately 100 year old (1999) beech trees grow on a benzothiazole (3.3), Na-metabilsulfite (95), EDTA-Na, moderate, south easternly inclined, slope at 465 to 515 (1 86), MgC1,-6H,0 (102), NADP (39.2), NAD (35.8), m above sea level on a sandy loam above quartzite. The sucrose (14 OOO), bovine serum albumine (500) and moderately well drained site was limed in 1989 with 3 TweenR 80 (500) (FIEDLER & ROTHE 1999). The tons per hectar of magnesium rich limestone (50 % homogenized sample was centrifuged for 25 min at 4 CaCO,, 40% MgCO,). The last selective logging was "C and 5 000 rpm (Minifuge GL, Heraeus Christ, performed in winter 1996. Osterode). Aliquots of 20 pl of the supernatand were transferred into 0.5 ml Eppendorf-tubes and stored at Root sampling -30 "C.

In the beech wood of forest district Heppenheim, a total Cellulose acetate electrophoresis of 61 root samples (individuals) were taken from the end of April to the beginning of June 1998 and ana- Electrophoresis was performed on 76 x 76 mm cellu- lyzed for isozyme patterns. In the beech wood of forest lose acetate plates (TitanRIII, Helena Laboratories, district Bingen, a total of 54 root samples (individuals) Helena Diagnostika, Hartheim), as described earlier were taken between mid of September and end of (FIEDLER & ROTHE 1999), but using the sample October 1998 and analyzed thereafter. About 1.5 to 2 m amounts and buffer systems given in Table 1. apart from the base of a beech, a soil square of about 1 m2 was freed from the litter. The upper soil was loos- Enzyme visualization ened by hand or using a spatula and investigated for the occurrence of fine roots. Then, fine roots were dug out After electrophoresis cellulose acetate plates were to a soil depth of approximately 5 cm, freed from placed, mylar side down, on a glass plate. A volume of coarse soil particles and inspected for white mycor- 2 ml of melted agar (20 mgm-', 60 "C) was added to rhizas of the type beech-Xerocomus chrysenteron. If the staining solution and the mixture (4 to 5 ml per root tips were infected with this fungus, the fine root (d plate) poured over the gel plates. After the agar had set, < 2 mm) to which it belonged, was followed until free the plates were incubated in the dark at 37 "C until of this and any other fungus. Each root sample was put enzyme bands were visible. into a separate plastic bag, numbered and stored on ice. Diaphorase (Cytochrome b5 reductase, EC 1.6.2.2) Within the next four days, the roots were transferred to was stained with a mixture of 1 ml of 100 mM Tris- a Petri dish with ice water, cleaned with a fine brush HCl, pH 8.5, 1.5 ml NADH (3 mgm-I), 5 drops of and finely selected for infections with X. chrysenteron dichlorophenolindophenole (DCPIP) (3 mgm-') and 5 and, identified under a stereo microscope at a 25-fold drops of MTT (3-[4,5-dimethylthiazol-2ylI-2,5- Table 1. Number of extract application and buffer systems and alleles at gene loci according to known enzymatic used to separate enzymes of Xerocolnus chryserzteron in structures of diploid material (ZHUet al. 1988; SOLTIS cellulose acetate electrophoresis. & SOLTIS1990; ROTHE1994). Definition and utilization of mathematical formulae were as proposed by Number of WRIGHT(1978), ROTHE(1994) and YEH et al. (1998). Enzyme system extract Buffer system The software POPGENE Vcrsion 1.31 (YEH et al. applications* 1998) with the following definitions and formulae was applied: genetic identity = I, genetic distances = D and Acid phosphatase 2 G d, (Table 9), heterozygosity = H, and HE (Table lo), Diaphorase 2 L Glucose-6-phosphate 2 T-C genetic diversities = H,, H,, D,, and G,, (Table 1 I), dehydrogenase fixation index = F, (Table 12). Differences in relative Leucine aminopeptidase 1 L frequencies (p %) of genotypes at a significance level

Malate dehydrogenase 2 L of a = 5 % were calculated by using the formula a @ =, Malic enzyme 4 L ,, = 1.96~with v,, = +.(@ (100 - p) 1 N)", with N: Peptidase 1 1 T-G number of samples per population (B. Thiebaut, per- (Gly-Leu as substrate) sonal communication). Peptidase 2 1 T-G (Leu-Gly-Gly as substrate) RESULTS *' one application corresponds to approximately 0.25 p1 Investigated enzyme loci Buffer G: Tris-maleic acid, pH 7.8 (RICHARIISONet al. 1986): 100 mM Tris, 40 mM nialelc acid Buffer L: Tris-maleic acid EDTA-MgC12, pH 7.8 Isoenzyme banding patterns were interpreted assuming (RICHARDSONet nl. 1986): 50 mM tris, 20 n&I maleic acid, that the mycelia of Xeroconzus chrysenteron are 1 mM EDTA- 1 mM Mg CI,-6H20 dikaryotic because in the ectomycorrhiza the basidio- Buffer T-C: Tris citrate, pH 7.0 (SOLTIS& SOLTIS1990): mycete forms a clamp mycelium (BRAND1991). The 135 mM Tris, 43 mM citric acid-monohydrate clamp mycelium represents a dikaryotic state which is Buffer T-G: Tris-glycine, pH 8.5 (HEUERT& BEATON genetically equivalent of a diploid (ESSER 1986). 1993): 25 mM Tris, 192 mM glycine (diluted 1 in 10 for use) Therefore genotype and allele frequencies at nine different enzyme gene loci (Table 2) were interpreted as diphenyltetrazolium bromide (10 mgm-') (cf. HARRIS for diploid organisms. Acid phosphatase, leucine & HOPKINSON1976) (Figure 2). aminopeptidase, malic enzyme and peptidase 2 are Peptidase 1 (EC 3.4.1 1.*) was stained with 2 rnl 20 assumed to be monomeric enzymes, whereas malate rnM phosphate buffer, pH 7.5, 4 drops of peroxidase dehydrogenase, diaphorase and peptidase 1 are thought (200 Units.mg-', 5 mgm-I; Boehringer, Mannheim), 8 to be dimeric enzymes (cf. ZHUet al. 1988) (Figures drops of o-dianisidine-2HC1 (4 mgm-'), 2 drops of 1-8). As an example of a monomeric enzyme with two MgC1,-6H,O (20 mgm-I), 8 drops of the dipeptide alleles, glucose-6-phosphate dehydrogenase can be glycine-leucine (5 mgm-') and 4 drops of amino acid taken (Figure 3), while an example of a dimeric enzyme oxidase (0.55 units.mg-'; 20 mgm-I; Sigma, Miinchen) with two alleles is peptidase 1 (Figure 7). The mono- (HEBERT& BEATON1993) (Figure 7). meric enzyme system shows two allelic variants when Peptidase 2 (EC 3.4.1 1.") was stained as for pepti- heterozygous and, one of the two when homozygous dase 1 but using 8 drops of the tripepetide leucine- (Figure 3) while the dimeric system shows three elec- glycine-glycine (5 mg . ml-') instead of the dipeptide trophoretic variants when heterozygous and one, either (HEBERT& BEATON1993) (Figure 8). the slowest or the fastest migrating form, when homo- Acid phosphatase (EC 3.1.3.2) (Figure I), glucose- zygous (Figure 7). 6-phosphate dehydrogenase (EC 1.1.1.49) (Figure 3), leucine aminopeptidase (EC 3.4.11.1) (Figure 4), Population genetic measures malate dehydrogenase (EC 1.1.1.37) (Figure 5) and malic enzyme (EC 1.1.1.40) (Figure 6) were visualized Polymorphic loci and effective number of alleles as previously described (FIEDLER& ROTHE1999). The nine polymorphic gene loci investigated showed a Exploitation of electrophoretic data total of 23 alleles, but two of these were found only in one of the investigated beech stands (Heppenheim), Because of the dikaryotic state of the hyphae, electro- namely the alleles Dia-B, and Me-B, (Table 3). There- phoretic enzyme patterns were translated into genotypes fore, the average number of alleles per polymorphic

O ARBORA PUBLISHERS C. VON MELTZER& G. ROTHE:GENETIC VARIATION OF XEROCOMUSCHRYSENTERON

Figure 1. Zymogram of acid phosphatase allozymes of Figure 2. Zymogramof diaphorase allozymes of mycorrhizae mycorrhizae of Fagus sylvaticaLYerocomus chrysenteron of Fagus sylvaticaLYerocornus chrysenteron from a beech from a beech stand in the forest district of Heppenheim. From stand in the forest district of Heppenheim. From left to right: left to right: Acp-B genotypes 111,212, control, 212, 112, 111, Dia-B genotypes: 212,213,111, control, 213,213,212,212,313, 111, 112,212,212. Control: Extract of non-mycorrhizal beech 213 and Dia-C genotypes: 111, 111, 111, control, 111, 112, 111, fine roots (1 to 2 mm in diameter). 111,212, 111. Control: Extract of non-mycorrhizal beech fine roots (1 to 2 mm in diameter).

Figure 3. Zymogram of glucose-6-phosphate dehydrogenase Figure 4. Zymogram of leucine aminopeptidase allozymes of allozymes of mycorrhizae of Fagus sylvaticalXerocomus mycorrhizae of Fagus sylvaticaLYerocomus chrysenteron chrysenteron taken from two beech stands in the forest taken from two beech stands in the forest districts of Bingen districts of Bingen (B) and Heppenheim (H). From left to (B) and Heppenheim (H). From left to right: Lap-B genotypes right: G6pdh-A genotypes 212 (H), 111 (B), 112 (H), control, 111 (B), 313 (B), 414 (B), 113 (H), 213 (B), control, 314 (H), 212 (B), 212 (H), 111 (B), 212 (B), 112 (H), 212 (H). Control: 313 (B), 113 (H), 313 (B). Control: Extract of non-mycorrhizal Extract of non-mycorrhizal beech fine roots (1 to 2 mm in beech fine roots (1 to 2 rnm in diameter) from Bingen. diameter) from Heppenheim.. Figure 5. Zymogram of malate dehydrogenase allozymes of Figure 6. Zymogram of malic enzyme allozymes of mycorrhizae of Fagus sylvatica/Xerocomus chrysenteron mycorrhizae of Fagus sylvaticdXerocomus chrysenteron from a beech stand in the forest district of Heppenheim. From from a beech stand in the forest district of Heppenheim. From left to right: Mdh-Cgenotypes 313,212, 111,313,213, control, left to right: Me-B genotypes 213, 213, 212, 212, 313, control, 212, 112, 313, 111. Control: Extract of non-mycorrhizal beech 213,313, 213, 113. Control: Extract of non-mycorrhizal beech fine roots (1 to 2 mm in diameter). fine roots (1 to 2 mm in diameter).

Figure 7. Zymogram of peptidase 1 allozymes (substrate: Figure 8: Zymogram of peptidase 2 allozymes (substrate: glycine-leucine) of mycorrhizae of Fagus sylvaticdXero- leucine-glycine-glycine) of mycorrhizae of Fagus sylvatica contus chrysenteron from a beech stand in the forest district flerocomus chrysenteron from a beech stand in the forest of Heppenheim. From left to right: Pep 1-B genotypes 112, district of Heppenheim. From left to right: Pep 2-B genotypes 111, 212,212, 111, 112,212, 112, control, 112. Control: Extract 112, 212, 112, 212, 112, control, 111, 212, 112, 212. Control: of non-mycorrhizal beech tine roots (1 to 2 mm in diameter). Extract of non-mycorrhizal beech fine roots (1 to 2 mm in diameter).

O ARBORA PUBLISHERS C. VON MELTZER& G. ROTHE:GENETIC VARIATION OF XEROCOMUSCHRYSENTERON

Table 2. Enzyme systems, corresponding enzyme loci, scored enyzme loci, number of subunits per enyzme and alleles per enyzme locus investigated in Xerocomus chrysenteron.

Enzyme system Loci Scored loci Subunits per enzyme Alleles per locus Acid phosphatase Diaphorase

Glucose-6-phosphate dehydrogenase Leucine aminopeptidase Malate dehydrogenase Malic enzyme Peptidase 1 Peptidase 2

Table 3. Number of alleles and effective number of alleles (n)of the nine enzyme loci investigated in the two Xerocomus chrysenteron populations Bingen and Heppenheim.

Bingen Heppenheim Enzyme locus Alleles; (number of Efective number of Alleles, (number of Effective number of alleles) alleles (n) alleles) alleles (n)

Acp-B Dia-B Dia-C G6pdh-A Lap-B Mdh-B Me-B Pep I -B Pep2-B

Mean value 2.33 1.53 2.55 1.56 locus was 2.55 for the X. chrysenteron population from and Mdh-C were found only in the Heppenheim Heppenheim and 2.33 for the population from Bingen. population or in the Bingen population. The effective number of alleles, n,, which increases Acid phosphatase showed three genotypes at locus linearly with the number of alleles and which may be Acp-B (Figure 1). The homozygous genotype 111 was used to quantify genetic variation between populations significant more frequent in the Bingen population was calculated to be n, = 1.56 for the Heppenheim (75.93 %; Heppenheim: 37.70 %) while the homozy- population, while the effective number of alleles for the gous genotype 212 was significant more frequent in the Bingen population was estimated to be n, = 1.53. Heppenheim population (45.90 %; Bingen: 12.96 %). The genotype 112 was equally frequent in both popula- Genotype frequencies tions (1 1.11 % and 16.39 %) (Tables 4 and 5). At the gene locus Dia-B four genotypes were The nine polymorphic gene loci investigated showed a observed (Figure 2). Of these, the type 212 was most total of 34 genotypes in both populations. Therefore, frequent in both populations (74.07 % and 75.41 %) the average number of genotypes per polymorphic locus (Tables 4 and 5). The genotype 111 was only observed was 3.22 for the X. chrysenteron population from in the Heppenheim population (1.64 %) (Table 5). The Bingen and 3.66 for the population from Heppenheim. genotypes 213 (9.26 % and 19.67 %) and 313 (16.67 % Several genotypes at locus Din-B, G6pdh-A, Lap-B and 3.28 %) were not significantly different in the Table 4. Relative frequencies (%) genotypes in the Xerocomus chrysenteron population from Bingen.

Genotype frequencies (%) at locus Genotype Acp-B Dia-B Dia-C G6pdh-A Lap-B Mdh-C Me-B Pepl-B Pep2-B

111 75.93l 0 83.33 3.70 3.70 3.70 0 16.67 14.81 112 11.11 0 7.41 0 0 0 0 37.04 62.96 113 - 0 - - 0 0 0 - - 114 - - - - 0 - - - 22.22 212 12.96' 74.07 9.26 96.30 0 72.22 12.96 46.30 - 213 - 9.26 - - 5.56 16.67 72.22 - - 214 - - - - 0 - - - - 313 - 16.67 - - 74.07 7.41 14.81 - - 314 - - - - 12.96 - - - - 414 - - - - 3.70 - - - -

0: Genotype not detected, no number: genotype not existing; I' frequency significantly different from that at stand Heppenheim at a significance level of a = 5%, investigated samples N = 54.

Table 5. Relative frequencies (%) of genotypes in the Xerocomus chrysenteron population from Heppenheim

Genotype frequencies (%) at locus Genotype Acp-B Dia-B Dia-C G6pdh-A Lap-B Mdh-C Me-B Pepl-B Pep2-B

111 37.70' 1.64 80.33 0 0 1.64 1.64 6.56 1 1.48 112 16.39 0 8.20 11.48 0 3.28 1.64 45.90 63.93 113 - 0 - - 6.56 0 8.20 ------114 - 0 - 212 45.90' 75.41 11.48 88.52 0 73.77 3.28 47S4- 24.95 213 - 19.67 - - 4.92 13.11 62.30 - 214 - - - - 0 - - - 313 - 3.28 - - 77.05 8.20 22.95 - - 314 - - - - 11.48 - - - - 414 - - - - 0 - - - -

0: Genotype not detected, no number: genotype not existing; I' frequency significantly different from that at stand Heppenheim at a significance level of a = 5%, investigated samples N = 61.

Bingen and Heppenheim population (Tables 4 and 5). (Figure 4). Of these, the type 313 was most prominent At locus Dia-C there were three genotypes (Figure in both populations (74.07 % and 77.05 %). The 2). The most prominent one in both populations was genotypes 111 (3.70 %) and 414 (3.70 %) appeared only type 111 (80.33 % and 83.33 %); the two remaining in the Bingen population, while the type 113 existed types 112 and 212 were equally frequent in both popula- only in the Heppenheim population (6.56 %). The tions (112: 7.41 % and 8.02 %; 212: 9.26 % and 11.48 genotypes 213 (5.56 % and 4.92 %) and 314 (12.96 % %) (Tables 4 and 5). and 11.48 %) were equally frequent in both populations Locus G6pdh-A showed three genotypes (Figure 3). (Tables 4 and 5). The type 212 was most frequent in both populations Locus Mdh-C comprises five genotypes (Figure 5). (96.30 % and 88.52 %, respectively) (Tables 4 and 5). Of these, the type 212 was most prominent in both Genotype 111 was exclusively found in the population populations (72.22 % and 73.77 %). The type 112 fromBingen (3.70 %), while the genotype 112 appeared exclusively appeared in the Heppenheim population only in the population from Heppenheim (1 1.48 %) (3.28 %). The genotypes 111 (3.70 %), 213 (16.67 %) (Tables 4 and 5). and 313 (7.41 %) were equally frequent in the Bingen At locus Lap-B six genotypes were observed population and the Heppenheim population (1.64 %,

O ARBORA PUBLISHERS 13.11 % and 8.20 %) (Tables 4 and 5). whereas allele 2 occurred at a lower frequency in the At locus Me-B six genotypes were observed. Figure Bingen population (18.52 %) than in the Heppenheim 6 shows four of them. The most prominent one was population (54.10 %) (Tables 6 and 7). All other type 213, in both populations (72.22 %, 62.30 % respec- investigated alleles appeared at nearly equal frequencies tively). The genotypes 111 (1.64 %), 112 (1.64 %) and in the two populations in Central Germany. 113 (8.20 %) were limited to the stand in Heppenheim. In the Bingen stand the types 212 (12.96 %) and 313 Hardy-Weinberg distribution of genotypes (14.81 %) were not significantly different from the Heppenheim stand (3.28 % and 22.95 %) (Tables 4 and The number of observed genotypes at the nine polymor- 5). phic enzyme loci corresponds at three loci (G6pdh-A, At locus Pepl-B three genotypes were seen (Figure Lap-B, Pepl-B) in both populations to a Hardy-Wein- 7). Type 111 (16.67 % and 6.56 %), type 112 (37.04 % berg distribution (Table 8). At four loci (Acp-B, Dia-C, and 45.90 %) and type 212 (46.30 % and 47.54 %) were Mdh-B, Me-B), however, significant deviations were equally frequent in both populations (Tables 4 and 5). observed in the populations of Bingen and Heppenheim At locus Pep2-B three genotypes also appeared (Table 8). At two gene loci, genotypes were distributed (Fig. 8). These were equally frequent in both popula- according to Hardy-Weinberg in one population tions (Tables 4 and 5). (Dia-B: Heppenheim, Pep2-B: Bingen) but not in the other one (Table 8). Allele frequencies Genetic identity I and genetic distances D and do Of the 23 alleles detected at the nine polymorphic enzyme loci, two were population specific. The alleles The genetic identity I between the two populations Dia-B, and Me-B, did not occur in the Bingen popula- Bingen and Heppenheim as defined by NEI (1972) was tion, whereas they were observed at low frequencies in calculated to be I = 97.28 % (Table 9). Correspond- the Heppenheim population (Me-B,: 6.56 %, Dia-B,: ingly, the genetic distance according to NEI(1972) is D 1.64 %; Tables 6 and 7). Larger allele frequency = 2.98 % and that of GREGORNS(1974) is do = 8.60 % differences were observed at locus Acp-B where allele (Table 9). The smallest genetic distance was observed 1 was more prominent in the Bingen population (8 1.48 at locus Mdh-B (D = 0.01 %, do = 1.41 %), the largest %) than in the Heppenheim population (45.90 %), locus Acp-B (D = 22.33 %, do = 35.58 %) (Table 9).

Table 6. Relative allele frequencies (5%) of the nine scored enzyme loci of the Xerocomus chrysenteron population from Bingen.

Allele Allele frequencies (%) at locus Acp-B Dia-B Dia-C C6pdh-A Lap-B Mdh-C Me-B Pepl-B Pep2-B

Table 7. Relative allele frequencies (%) of the nine scored enzyme loci of the Xerocomus chrysenteron population from Heppenheim.

Allele Allele frequencies (%) at locus Acp-B Dia-B Dia-C G6pdh-A Lap-B Mdh-C Me-B Pepl-B Pep2-B Table 8. Chi-squared (xZ) values, probability values (P) allocated to the values and significance levels (S) at which the number of genotypes deviate froma Hardy-Weinberg distributionin theX. chrysenteron population fromBingen and from Heppenheim.

Acp-B Dia-B Dia-C Bingen Heppenheim Bingen Heppenheim Bingen Heppenheim

x2 16.3281 25.5127 35.3737 3.3766 17.2363 37.8441 P 5.33*10-5 4.39*10-7 l.01*10-~ 0.3371 3.3*10-5 7.66*10-'O S *h* *** *** *** ***

G6pdh-A Lap5 Mdh-B Bingen Heppenheim Bingen Heppenheim Bingen Heppenheim

x2 3.1075 0.0403 4.69 0.19 15.24 20.99 P 0.0779 0.8408 0.5845 0.9999 0.00162 0.000106 S *** ***

Bingen Heppenheim Bingen Heppenheim Bingen Heppenheim

Table 9. Genetic identity and genetic distances of the two Table 10. Heterozygosities (H,and HE)of the nine enzyme Xerocomus chrysenteron populations Bingen and Heppen- loci investigated in the Xerocomus chrysenteron heim. populations from Bingen and from Heppen-heim.

Enzyme Genetic Genetic Genetic Enzyme locus HLl Bingen HL l Heppenheim locus identity 1 distance D distance do 0.3018 0.4966 0.3353 0.2558 0.2256 0.2629 0.07 13 0.1082 0.2931 0.21 14 0.3249 0.3053 0.4998 0.5327 0.4561 0.4160 0.4973 0.49 14

Total 0.9782 0.0298 0.0860 was observed at locus G6pdh-A (Bingen: HL= 7.13 %, Heppenheim: H, = 19.82 %) (Table lo), while the Heterozygosity largest heterozygosity was found at locus Me-B (Bin- gen: H, = 49.98 %, Heppenheim: H, = 53.27 %) (Table The average heterozygosity over all investigated gene 10). loci, HE, was similar in both investigated X. chrysen- teronpopulations (Bingen: HE=34.23 %; Heppenheim: Genetic diversity HE = 33.39 %) (Table 10) which means that their genetic diversity is very similar. The amount of The total amount of genetic diversity, H, (NEI 1972) heterozygosity per locus, however, varied in both between bothX. chrysenteron populations was found to populations considerably. The lowest heterozygosity be H, = 34.66 (Table 11). The within population

O ARBORA PUBLISHERS C. VON MELTZER& G. ROTHE:GENETIC VARIATION OF XEROCOMUSCHRYSENTERON genetic diversity, H,,was calculated to be H, = 33.81 % Table 12. Fixation index (F,) of the nine investigated while the genetic diversity between both populations, enzyme loci of the Xerocomus chrysenteron populations D,,, was estimated to be D,, = 0.85 % (Table 11). The from Bingen and from Heppenheim as well as the total fixation index of both populations (F,,). per cent amount of D,, of the total genetic diversity (H,), G,,, was estimated to be 2.45 % (Table 11). At locus Mdh-B the total genetic diversity was found to be F,s lowest within the populations (D,, = 0) while the Enzyme locus genetic diversity between the two populations was Bingen Heppenheim Total largest at locus Acp-B (G,, = 13.69 %) (Table 11).

Table 11. Genetic diversities H, H,, D,, and G,,.

Enzyme locus

Acp-B Dia-B Dia-C G6pdh-A Lap-B per polymorphic locus (A/L) is relatively high (X. Mdh-B chrysenteron: AIL = 2.56; F. sylvatica: AL = 2.3, Me-B MULLER-STARCK199 1). The genetic variability of both Pepl-B species is much higher within populations than between Pep2-B Mean value populations (X. chrysenteron: average total genetic distance, d ,,,,, = 8.6 %; F. sylvatica: d = 5.3 % (SANDERet al., submitted). Fixation Index The genetic diversity (NEI 1973) of both symbiotic partners is similar with respect to total genetic diversity The fixation index (WRIGHT1978) was negative in both (H,), average genetic diversity within (H,) and between populations at the loci Me-B (F,,,,,,, = -0.3762) and populations (D,,), and percent amount of D,, on H, Pep2-B (F,,,,,,, = -0.2835) (Tablel2) which indicates (G,,) (X. chrysenteron: H, = 0.347, H, = 0.338, DsT = an excess of heterozygous individuals. On the other 0.009, G,, = 0.025; F. sylvatica: H, = 0.296, Hs = hand, both populations differ with respect to the num- 0.293, D,,= 0.003, G,,= 0.009, SANDERet al., submit- ber of heterozygous individuals. In the Bingen popula- ted). tion, the number is lower than in the Heppenheim The average expected heterozygosity varies little population at the loci G6pdh-A , Lap-B and Pepl-B between populations of both symbiotic partners (X. (Table 12). chrysenteron: 0.338; F. sylvatica HE = 0.396, SANDER et al., submitted). On the other hand, the heterozygosity DISCUSSION of individual loci varies considerably (X. chrysenteron: H, = 0.10888 to 0.533; F. sylvatica: H, = 0.123 to Isozyme banding patterns were interpreted as if the 0.6798, SANDERet al., submitted). mycelia of X. chrysenteron were diploid because they Differences between host and symbiotic fungus grow in mycorrhizas with beech as a dikaryotic myce- were observed with respect to the occurrence of alleles lium (BRAND1989). However, the loci and alleles and genotypes in their respective populations. Whereas identified in the fungus are putative. Crosses of mono- no population specific alleles were found in European karyotic isolates are currently under study. It is also beech (BELLETTI& LANTERI1996; LEONARDI& assumed that individual samples from the very same MENOZZI1995; KONNERT1995; LOCHELT& FRANKE tree rootlet were small enough to contain only one 1995; HATTEMER& ZIEHE1996) populations specific fungal individual. DNA analyses of the IGS 1 region of alleles were observed in X. chrysenteron (Me-B: allele the rDNA gene cluster do not contradict to this view 1; Dia-B: allele 1). The effective number of alleles is (HAESE& ROTHE,in preparation). lower in the fungus (X. chrysenteron: n, = 1.55) than in Both partners of the symbiotic entity F. sylvatica/X. its host (F. sylvatica: n, = 2.67, SANDERet al., submit- chrysenteron show common and differing genetic ted). Populations of European beech may show differ- properties. Both possess, for example, the following ent genotype frequencies at several loci (SANDERet al., common characteristics: the average number of alleles submitted) but the two X. chrysenteron populations showed unique genotypes at several loci (Dia-B 111, GREGORIUS,H. R. 1974: Genetischer Abstand zwischen G6pdh-A 111 and 112, Lap-B 111, 113 and 414, Mdh-C Populationen. I. Zur Konzeption der genetischen 112, Me-B 111, 112 and 113). If the corresponding Abstandsmessung. Silvae Genetica 23 (1-3): 22-27. HARRIS,H. & HOPKINSON,D. A. 1976: Handbook of enzyme enzymes have different physicochemical properties or electrophoresis in human genetics, American Elsevier, are localized at different cell compartments, their New York, 4-1 pp. presence or absence would indicate metabolic differ- HATTEMER,H. H. & ZIEHE,M. 1996: An attempt to infer on ences in both species with respect to glycolysis the origin of a Beech (Fagus sylvatica L.) stand in (G6pdh), malic acid-metabolism (Mdh, Me), fatty acid- Rheinland-Pfalz (Germany). Silvae Genetica 45: 276-283. metabolism (Dia) and peptide cleavage (Lap). The soil HEBERT,P. D. N. & BEATON,M. J. 1993: Methodologies for at both stands is different and the two populations could allozyme analysis using cellulose acetate electrophoresis. have adapted to this situation. However, a limited gene A practical handbook. An educational service of Helena flow between both populations of X. chrysenteron is Laboratories, Beaumont, Texas, U. S. A, 32 pp. KONNERT,M. 1995: Investigations on the genetic variation of also possible due to the natural borders of the river beech (Fagus sylvatica L.) in Bavaria. Silvae Genetica 44: Rhine. 346-351. The common genetic properties of European beech LEONARDI,S. & MENOZZI,P. 1995: Genetic variability of and X. chrysenteron are most possibly the result of a Fagus sylvatica L. in Italy: the role of postglacial co-evolution which started after Fagus sylvatica recolonization. Heredity 75: 35-44. occupied its natural range after the last glaciation some LITTKE,W. R., BLEDSOE,C. S. & EDMONDS,R. L. 1984: 11 000 years ago. This opinion is not in contradiction Nitrogen uptake and growth in vitro by Hebeloma to the result that Canadian populations of Suillus crustuliniforme and other Northwest mycorrhizal fungi. tomentosus showed larger interpopulation genetic Can. J. Bot. 62: 647-652. LOCHELT,S. & FRANKE,A. 1995: Bestimmung der geneti- variabilities than intrapopulation genetic variabilities schen Konstitution von Buchen-Bestanden (Fagus when each population infected different species of trees sylvatica L.) entlang eines Hohentransektes von Freiburg (ZHUet al. 1988). auf den Schauinsland. Silvae Genetica 44: 3 12-3 18. LUNDEBERG,G. 1970: Utilization of various nitrogen sources, ACKNOWLEDGEMENT in particular bound soil nitrogen by mycorrhizal fungi. Stud. For. Suec. 79: 1-95. This work was supported financially by the Deutsche For- MARX,D.H. 1981: Variability in ectomycorrhizal develop- schungsgemeinschaft. ment and growth among isolates of Pisolithus tinctorius as affected by source, age, and reisolation. Can. J. For. REFERENCES Res. 11: 168-174. MOLINA,R. 1979: Ectomycorrhizal inoculations of containe- rized Douglas-fir and lodgepole pine seedlings with six AGERER, R. 1988-1998: Color Atlas of Ectomycorrhizae, isolates of Pisolithus tinctorius. For. Sci. 25: 585-590. Einhom-Verlag, Schwabisch-Gmiind, 140 pp. MULLER-STARCK,G. 1991: Survey of genetic variation as BELLETI,P. & LANTERI,S. 1996: Allozyme variation among inferred from enzyme gene marker. In: Genetic variation European beech (Fagus sylvatica L.) stands in Piedmont, in European populations of forest trees. (eds. G. Miiller- North-Westem Italy. Silvae Genetica 45: 33-37. Starck and M. Ziehe). pp 20-37. Sauerlander's Verlag, BRAND,F. 1989: Studies on ectomycorrhizae. XXI. Beech Frankfurt am Main. ectomycorrhizae and rhizomorphs of Xerocomus chrysen- MULLER-STARCK,G. & STARKE,R. 1993: Inheritance of teron (Boletales).Nova Hedwigia 48: 469-483. isoenzymes in European beech (Fagus sylvatica L.). BRAND,F. 1991: Ektomykorrhizen an Fagus sylvatica. Cha- Journal of Heredi~84: 291-296. rakterisierung und Identifizierung, okologische Kenn- NEI, M. 1972: Genetic distance between populations. The zeichnung und unsterile Kultivierung. IHW-Verlag, American Naturalist 106 (949): 283-292. Eching, 229 pp. NEI, M. 1973: Analysis of gene diversity in subdivided DAHNE,J., KLINGELHOFER,D., OTT, M. & ROTHE, G. M. populations. Proc. Acd. Sci. USA 70: 3321-3323. 1995: Liming induced stimulation of the amino acid RICHARDSON,B. J., BAVERSTOCK,P. R. & ADAMS,M. 1986: metabolismin mycorrhizal roots of Norway spruce (Picea Allozyme electrophoresis. A handbook for animal system- abies [L.] Karst.). Plant and Soil 173: 67-77. atics and population studies, Academic Press, Sydney, ESSER, K. 1986: Kryptogamen, Springer-Verlag, Berlin, Australia, 410 pp. Heidelberg, Germany, 566 pp. ROTHE,G. M. 1994: Electrophoresis of enzymes, Springer FIEDLER,0. & ROTHE,G. M. 1999: Cellulose acetate electro- Verlag, Berlin, Heidelberg, Germany, 307 pp. phoresis of the beech ectomycorrhiza Fagus sylvatica - SANDER,T., ROTHE,G. M., WEISGERBER,H. & JANBEN, A. Xerocomus chqsenteron. In: From Symbiosis to Euka- 1999: Analyse der genetischen Variabilitat und Diversitat ryotism, Endocytobiology VII. (eds. E. Wagner, J. Nor- von Rotbuchen (Fagus sylvatica L.) - Bestanden in man, H. Greppin, J. H. P. Hackstein, R. G. Herrmann, K. Hessen. Submitted for publication. V. Kowallik, H. E. A. Schenk, J. Seckbach). Vol 1. SOLTIS,D. E. & SOLTIS,P. S. 1990: Isozymes in Plant Biol- University of Geneva, 1999, pp. 253- 269. ogy. Chapman and Hall Ltd., London, UK, 268 pp.

O ARBORA PUBLISHERS 95 C. VON MELTZER& G. ROTHE:GENETIC VARIATION OFXEROCOMUS CHRYSENTERON

WRIGHT,S. 1978: Variability within and among Natural ZHU,H., HIGGINBOTHAM,K. O., DANCIK,B. P. & NAVRATIL, Populations, Vol. 4, The University of Chicago Press, S. 1988: Intraspecific genetic variability of isozymes in Chicago, USA, 580 pp. the ectomycorrhizal fungus Suillus tomentosus. Canadian YEH,F. C., YANG,R. C. &BoYLE,T.1998: PopGene Version Journal of Botany 66: 588-594. 1.31. Microsoft Window-based freeware for population genetic analysis.